We trust services like Last.fm or Pandora to learn our musical tastes and serve up custom radio stations, so why not the same for the numerous streams that bombard us daily?
My6Sense, a recommendation engine for your social streams and news feeds, is releasing its functionality today in its “Attention API” at the DEMO conference.
The company released an iPhone app last summer, which uses the company’s “digital intuition” to determine the particular stories and status updates that are most relevant to you according to what you’ve clicked on and spent time with in the past. The system learns from you, becoming more familiar with what you are interested in as time goes on.
According to the company, “the service is content/stream agnostic and can automatically rank information from all types of sources — including social streams like Twitter, news streams, RSS, vertical content providers, open whiteboards and more. The solution is optimized for mobile platforms, where digital clutter is the most prominent, and can be successfully used on non-mobile sites and applications as well.”
With services like Pandora, you need to let the program know when you like or dislike something. The “Attention API”, on the other hand, simply requires that the site making use of it report back user activity in order to learn the user’s habits and preferences.
While this sounds great, and we can see it filtering out some of the noise, we have the same concerns about it as we do about Google customizing our search – that we will end up in an echo chamber of like-minded thought. As our own Frederic Lardinois pointed out when reviewing the My6Sense iPhone app, you may want to step outside the recommendations once in a while for a breath of fresh air.
If you are a real news junkie, you will probably still sometimes want to switch to the regular timeline mode that organizes items chronologically. After all, the items you don’t usually think you would be interested in can sometimes really grab your attention (which is, to be honest, a problem that all recommendation systems have to grapple with).
On the company’s API page, it mentions that the API will provide “No fear of bombardment” as “Developers and publishers can broadcast any amount of information and content to a widespread audience, which reaching individual consumers with messages that are uniquely relevant to them.”
We would hope that, in reality, any service using the API would notify its users that it was doing so and even offer the ability to step outside of the service’s recommendations.
